N-Substituted amino acid N′-benzylamides: Synthesis, anticonvulsant, and metabolic activities

Article abstract of DOI:10.1016/j.bmc.2004.02.043

Amino acid amides (AAA) were prepared and evaluated in seizure models. The AAA displayed moderate-to-excellent activity in the maximal electroshock seizure (MES) test and were devoid of activity in the subcutaneous Metrazol-induced (scMet) seizure test. The AAA anticonvulsant activity was neither strongly influenced by the C(2) substituent nor by the degree of terminal amine substitution. An in vitro metabolism study suggested that the structure-activity relationship pattern was due, in part, to metabolic processes that occurred at the N-terminal amine unit.

Full text of DOI:10.1016/j.bmc.2004.02.043

Bioorganic & Medicinal Chemistry 12 (2004) 3079–3096
N-Substituted amino acid N0-benzylamides: synthesis,
anticonvulsant, and metabolic activities
a
a
b
c
ꢀ
ꢀ
Cecile Beguin, Arnaud LeTiran, James P. Stables, Robert D. Voyksner and
Harold Kohnd,*
aDepartment of Chemistry, University of Houston, Houston, TX 77204-5641, USA
bEpilepsy Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, 6001 Executive Boulevard,
Suite 2106, Rockville, MD 20852, USA
cLCMS Limited, PO Box 27228, Raleigh, NC 27611-7228, USA
dDivision of Medicinal Chemistry and Natural Products, School of Pharmacy, CB #7360, University of North Carolina, Chapel Hill,
NC 27599-7360, USA
Received 13 November 2003; accepted 11 February 2004
Available online 20 April 2004
Abstract—Amino acid amides (AAA) were prepared and evaluated in seizure models. The AAA displayed moderate-to-excellent
activity in the maximal electroshock seizure (MES) test and were devoid of activity in the subcutaneous Metrazol-induced (scMet)
seizure test. The AAA anticonvulsant activity was neither strongly influenced by the C(2) substituent nor by the degree of terminal
amine substitution. An in vitro metabolism study suggested that the structure–activity relationship pattern was due, in part, to
metabolic processes that occurred at the N-terminal amine unit.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
(CH3C(O)N(H)) unit in 1 was replaced with methyl,
methoxy, hydroxy, acetoxy, or halogen we obtained
compounds with diminished anticonvulsant activity.4;5
Amino acids that are functionalized at both the N- and
C-termini are proven potent anticonvulsant agents.1–3
These compounds are termed functionalized amino
acids (FAA, 1). We conducted a SAR study of >250
compounds1 that identified (R)-N-benzyl-2-acetamido-
3-methoxypropionamide ((R)-2) as the lead compound.1j
(R)-2 has entered phase II clinical trials for the treat-
ment of epilepsy and neuropathic pain under Schwarz
Pharma sponsorship. An important FAA structural
unit is the N-terminus. In the initial design of FAA, the
N-terminal amine was protected as an amide to provide
compounds with increased lipophilicity.1a Subsequent
studies demonstrated the importance of the acetamido
unit (R1 ¼ C(O)CH3) for potent anticonvulsant activity
and showed that either a decrease1k (i.e., R1 ¼ C(O)H)
or increases1b (i.e., R1 ¼ C(O)CH2CH3, C(O)C(H)-
(CH3)2, C(O)C(CH3)3) in the size of this moiety led to
reduced activity. Furthermore, when the acetamido
The importance of the intact acetamido group in FAA
has recently been reexamined by our1k;6 and Paruszew-
ski’s2a;b;d research groups. Both reported that substituted
amino acid derivatives, in which the acetyl (C(O)CH3)
substructure within the acetamido (CH3C(O)N(H))
moiety in FAA was replaced with either a hydrogen or a
small alkyl unit to give primary and secondary amines,
respectively, prevented seizures in the animal models.
The finding for N,N-dialkyl amino acid derivatives was
similar. Significantly, conversion of the acetamido unit
in 1 to an amino moiety provided amino acid amides
(AAA, 3) that are likely to have increased water solu-
bility compared with their FAA (1) counterparts. The
hydrophilicity of the terminal amino unit may permit
Keywords: Amino acid N0-benzylamides; Synthesis; Anticonvulsants;
Metabolism.
* Corresponding author. Tel.: +1-919-966-2680; fax: +1-919-843-7835;
e-mail: harold_kohn@unc.edu
0968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.02.043

3080
C. Bꢀeguin et al. / Bioorg. Med. Chem. 12 (2004) 3079–3096
the introduction of lipophilic groups at the other sites of
the molecule while still retaining the necessary physio-
chemical properties for drug administration. In this
study, we present an in-depth inspection of the N-ter-
minus requirements within substituted AAA (3) for
anticonvulsant activity, demonstrate that AAA are
potent anticonvulsants, and document that the terminal
amino group is prone to metabolic change.
31,1b 32,1b 2,1j served as our reference compounds. For
these FAA, we found a significant improvement in
anticonvulsant activities in maximal electroshock-
induced seizure test (MES) in mice and rats as we pro-
gressed from 31 to 32 to 2.
For most compounds, we prepared the racemic AAA,
since we did not know if either isomer would exhibit
preferential activity. This choice obviated the need to
prepare optically pure amines. To test if AAA, like
FAA, exhibited a stereochemical preference for anti-
convulsant activity1c–e;g;j;l we synthesized the individual
(R)- or (S)-enantiomers or both for 4, 24, and 26.
The anticonvulsant activities for AAA 4–30 were eval-
uated in rodents at the NIH’s Anticonvulsant Screening
Program (ASP) (Rockville, MD). In one of the initial
tests, the compound was administered to mice intra-
peritoneally (ip), followed by the determination of its
anticonvulsant properties measured by means of the
MES and the subcutaneous Metrazol-induced convul-
sion (scMet) models.7;8 In the second test, the compound
was administered to rats orally (po) and then evaluated
in the MES-seizure model. Neurological toxicity (Tox)
was also determined. For mice it was accomplished
using the rotorod model,9 and in rats by the observation
of motor impairment utilizing assessment of muscle
tone, positional sense, righting, and gait and stance.
Selected compounds were further evaluated in the MES
test in mice (po) and rats (ip) under a more strict
monitoring of dosages and activity time spans.
2. Results and discussion
2.1. Choice ofcompounds and methods ofpharmacolog-
ical evaluation
AAA structural sites 1 and 2 in AAA (Fig. 1) were
systematically altered to provide 4–30 (Fig. 2). We var-
ied the AAA N-terminus (site 1) by incorporating acy-
clic (primary, secondary) and cyclic alkylamino, and
alkoxyamino units at this position. The second was the
R2 moiety (site 2). Here, we restricted our choice to the
methyl (CH3), phenyl (C6H5), or methoxymethyl
(CH2OCH3) units to give AAA series A–C, respectively.
All three units were evaluated in the FAA series, thus
2.2. Synthesis
The AAA were prepared from the corresponding
2-halogen or 2-mesylate of the 2-substituted benzyl
amide 33 followed by either amine or hydroxylamine
displacement (Scheme 1). In series A, 2-chloropropionyl
chloride (34) and 2-bromopropionyl bromide (35) served
as our starting materials, and in series B, 2-chloro-2-
Figure 1. Sites of AAA modification.
Figure 2. AAA selected for study.

C. Bꢀeguin et al. / Bioorg. Med. Chem. 12 (2004) 3079–3096
3081
Scheme 3. Synthesis of (R)-N-benzyl-2-ethylamino-3-methoxyprop-
ionamide.
Scheme 1. General synthetic route to AAA.
2.3. Pharmacological evaluation
phenylacetyl chloride (36) was employed in most syn-
theses (Scheme 2). Treatment of 34–36 with an equiva-
lent of benzylamine gave 37,10 38, and 39, respectively.
We found it necessary to use mesylate 41 in place of 39
for reactions involving hydroxylamines. Mesylate 415
was prepared from alcohol 405 with mesyl chloride and
pyridine (Scheme 2). For series C AAA, we synthesized
45 from methyl acrylate (42) using the protocol of
Carter and West (Scheme 2).11 Intermediates 37–39, 41,
and 45 were then treated with commercially available
amines and hydroxylamines to afford compounds 5–13,
15–23, and 25–30 in 35–99% yields (Fig. 2). Racemic
amines 4,1a 6,11 9,2a 14,1b 15,12 23,13 24,10 and optically
active amines (R)-414 and (S)-415 have been reported.
Optically pure (R)-26 was synthesized in 51% yield by
condensation of amine (R)-24 with acetaldehyde fol-
lowed by imine reduction with sodium cyanoborohy-
dride (Scheme 3).
The in vivo anticonvulsant activities for racemic amines
4–30 are summarized in Tables 1 and 2, along with MES
activities of the FAA 31, 32, and 2. All compounds were
administered ip to mice (Table 1) and po to rats (Table
2). We have divided the tables into sections. Listed
horizontally are AAA that conform to the C(2) methyl
(alanine), the C(2) phenyl (phenylglycine), and the C(2)
methoxymethyl (O-methylserine) series (series A–C).
Running vertical are N-terminal amines listed in
increasing structural complexity. We begin with the
unsubstituted amines and progress to secondary and
tertiary amines. Compounds 4,1a 5,2a and 92a were pre-
viously tested in the MES seizure model in mice (ip) and
52a in rats (po). We retested 4 and 9 both in mice and
rats.
In mice (Table 1), we found that the ED50 values for the
C(2) methyl analogues (series A) were, typically, 30–
100 mg/kg, and only 4, 6, and 12 exhibited activities
greater than 100 mg/kg. The activity observed for C(2)
methyl AAA was comparable with that reported for the
parent FAA 31 (ED50 ¼ 76 mg/kg).1b We saw a wider
spectrum of activity for the C(2) phenyl AAA series. We
found compounds exhibiting excellent (ED50 < 30 mg/
kg) (16), moderate (ED50 ¼ 30–100 mg/kg) (15, 18, 19,
21, 22), and weak (ED50 ¼ 100–300 mg/kg) (14, 17, 20,
23) activity. Within this series, 15 (ED50 ¼ 46 mg/kg), 16
(ED50 < 30 mg/kg), and 19 (ED50 ¼ 36 mg/kg), possessed
anticonvulsant activity comparable with FAA 32
(ED50 ¼ 20 mg/kg)1b and phenobarbital (ED50 ¼ 22 mg/
kg).16;17 The final AAA series were the C(2) meth-
oxymethyl derivatives 24–30 (series C). The corre-
sponding FAA was 2, whose ED50 value was 8.3 mg/
kg.1j The primary (24) and secondary AAA (25–28)
displayed anticonvulsant activities from 30–100 mg/kg,
while the tertiary amines (19, 30) exhibited ED50 values
greater than 100 mg/kg. The neurological toxicities9 for
AAA in mice (ip) generally fell within the 100–300 mg/
kg range. We observed no appreciable anticonvulsant
activity (ED50 > 100 mg/kg) for any AAA in the scMet
test7;8 at doses up to 300 mg/kg for compounds evalu-
ated at 0.5 and 4 h (data not shown). A similar finding
has been reported for FAA 31, 32, and 2.1b;j
1
Spectral data (IR, H NMR, 13C NMR, low-resolution
mass, high-resolution mass) consistent with their pro-
posed structures were obtained for the compounds
prepared in this study. Satisfactory elemental analyses
were obtained for all compounds other than 28.
Inspection of the composite data in Table 2 revealed
that all AAA evaluated in rats (po) possessed significant
anticonvulsant activities, with ED50 values lower than
100 mg/kg. In the C(2) methyl series, the primary amine
4 displayed the highest potency (ED50 ¼ 14 mg/kg), three
times more potent than FAA 31.1b The activities for
secondary and tertiary C(2) methyl AAA 5–13 were 30–
100 mg/kg, except for 52a (ED50 ¼ 27 mg/kg). Similarly,
Scheme 2. Synthesis of N-benzyl-2-substituted amides.

Table 1. Selected pharmacological data for AAA and FAA in micea
Series A
Series B
MES,b
ED50
Series C
MES,b
ED50
Compd structure
MES,b ED50 Tox,c
TD50
PId
Compd structure
Tox,c
TD50
PId
4.8
Compd structure
Tox,c
TD50
PId
5.2
OCH3
O
76 [1]
(67–89)
454 [0.5]
(420–500)
5.9
20 [0.5]
(17–25)
97 [0.5]
(80–118)
8.3 [0.5]
(7.9–9.8)
43 [0.25]
(38–47)
H
N
N
Ph
H
O
32e
31e
2f
Ph
H
OCH3
H
CH3
H
>100, <300
>300
––
>100,
<300
>100,
<300
––
1.8
––
––
––
84 [0.25]
(65–97)
290 [0.25]
(240–320)
3.5
4.2
2.9
––
N
Ph
N
Ph
Ph
H2N
H2N
N
Ph
H2N
O
O
O
4
14
24
OCH3
CH3
Ph
H
H
31
(21–41)
99
(75–121)
3.2
46 [0.25]
(34–59)
83 [0.25]
(64–104)
68 [0.25]
(55–96)
290 [0.25]
(250–330)
H
N
Ph
N
N
H
N
Ph
N
H
N
H
O
O
O
5g
15
25
Ph
OCH3
CH3
H
Ph
H
N
N
>100
>100,
<300
––
<30
>30,
<100
51 [0.25]
(42–61)
150 [0.25]
(120–180)
H
Ph
N
N
Ph
N
H
H
O
N
H
O
O
16
6
26
OCH3
Ph
CH3
H
H
94 [0.25]
(76–125)
410 [0.5]
(290–590)
4.3
>100,
<300
>100,
<300
>30,
<100
>100,
<300
H
N
Ph
N
Ph
CH3O
CH3O
CH3O
N
Ph
N
H
N
CH3O
CH3O
N
H
H
O
O
O
7
17
27
Ph
OCH3
CH3
O
H
N
H
N
>30,
<100
>100,
<300
––
>30,
<100
>30,
<100
>30,
<100
>100,
<300
––
H
CH3O
Ph
Ph
N
H
N
Ph
N
H
N
H
O
O
8
18
28
CH3
Ph
OCH3
H
H
74 [0.25]
(61–90)
160 [0.25]
(140–180)
36 [0.25]
(30–46)
72 [0.25]
(57–86)
2.0
>300
ꢀ300
––
2.1
H
N
Ph
N
Ph
N
N
Ph
N
N
O
O
O
9g
19
29

Ph
CH3
H
N
H
N
>30,
<100
>100,
<300
––
––
>100,
<300
>100,
<300
––
––
––h
Ph
Ph
CH3O
CH3O
CH3O
CH3O
N
N
CH3
O
CH3
O
10
20
Ph
CH3
O
H
N
H
N
>30,
<100
>100,
<300
>30,
<100
>100,
<300
––h
Ph
Ph
N
N
O
CH3O
CH3
CH3O
Ph
11
21
H
N
H
N
>100, <300
>100,
<300
––
––
>30,
<100
>30,
<100
––
––
––h
Ph
Ph
Ph
N
N
O
O
22
12
CH3
Ph
OCH3
H
H
>30,
<100
ꢀ300
>100,
<300
>100,
<300
>100,
<300
ꢀ300
––
N
Ph
H
N
N
N
N
Ph
N
O
O
O
O
O
O
13
Phenytoini
23
30
9.5 [2]
(8.1–10)
22 [1]
66 [2]
6.9
(53–72)
69 [0.5]
(63–73)
430 [0.25]
(370–450)
Phenobarbitali
Valproatei
3.2
1.6
(15–23)
270 [0.25]
(250–340)
a The compounds were administered intraperitoneally.
b MES ¼ maximal electroshock seizure test in mg/kg. Numbers in parenthesis are 95% confidence intervals. The dose effect data was obtained at the ‘time of peak effect’ (indicated in hours in the brackets).
c Tox ¼ neurologic toxicity determined from rotorod test in mg/kg. Numbers in parenthesis are 95% confidence intervals. The dose effect data was obtained at the ‘time of peak effect’ (indicated in hours in
the brackets).
d PI ¼ protective index (TD50/ED50).
e Ref. 1b.
f Ref. 1j.
g Ref. 2a.
h Not prepared.
i Ref. 17.

Table 2. Selected pharmacological data for AAA and FAA in ratsa
Series A
Series B
MES,b
ED50
Series C
Compd structure
MES,b ED50 Tox,c
TD50
PId
Compd structure
Tox,c
TD50
PId
Compd structure
MES,b
ED50
Tox,c
TD50
PId
OCH3
O
48 [1]
(32–72)
>1000
>21
48 [4]
(31–68)
>1000
>21
3.8 [2]
(2.9–5.5)
390 [1]
(320–520)
102
H
N
N
Ph
H
O
32e
31e
2f
Ph
H
OCH3
H
CH3
H
14 [1]
>500
>36
>18
>30,
<100
>30
––
29 [1]
>450
>400
>16
>14
N
Ph
N
Ph
Ph
(7–22)
(18–45)
H2N
H2N
N
Ph
H2N
O
O
O
4
14
24
OCH3
CH3
Ph
27
>500
20 [0.25]
(8–39)
>63,
<500
>3.1
28 [2]
H
H
H
N
Ph
N
(16–48)
(19–40)
N
H
N
Ph
N
H
N
H
O
O
O
5g
15
25
Ph
OCH3
CH3
H
Ph
H
N
N
32 [4]
(19–46)
>125
>500
>3.9
20 [0.25]
(14–32)
>125,
<250
––
––
22 [0.25]
(13–33)
>500
>500
>23
>23
H
Ph
N
N
Ph
N
H
H
O
N
H
O
O
16
6
26
OCH3
Ph
CH3
H
H
37 [2]
>13.5
>30
>30
22 [0.5]
(14–29)
H
N
Ph
N
Ph
CH3O
CH3O
CH3O
N
Ph
N
H
N
CH3O
CH3O
N
N
(25–52)
H
O
O
H
O
7
17
27
Ph
OCH3
CH3
O
H
N
H
N
>4.5
62 [2]
>400
>500
>6.5
>5.7
28 [0.5]
(21–38)
>125
32 [0.5]
(20–44)
>250
>500
>7.8
H
CH3O
Ph
Ph
N
H
N
Ph
N
H
N
(38–98)
O
H
O
8
18
28
CH3
Ph
OCH3
H
N
H
N
88 [4]
7.1 [0.25] 230 [0.25]
(5.0–10) (190–260)
33
35 [0.25]
(29–42)
>14
H
Ph
Ph
N
N
Ph
N
(53–130)
O
O
O
9
19
29

Ph
CH3
H
N
H
N
ꢀ30
ꢀ30
>30
>30
––
––
>30,
<100
>30
>30
––
––
––h
Ph
Ph
CH3O
CH3O
CH3O
CH3O
N
N
CH3
O
CH3
O
10
20
Ph
CH3
O
H
N
H
N
>30,
<100
––h
Ph
Ph
N
N
O
CH3O
CH3O
11
21
CH3
Ph
H
N
H
N
<50
>50
––
<30
>30
>30
––
––
––h
Ph
Ph
Ph
N
N
O
O
12
22
CH3
Ph
OCH3
H
H
33 [2]
>250
>7.6
>30,
<100
>30,
<100
>300
––
N
Ph
H
N
N
N
N
Ph
(26–42)
N
O
O
O
O
O
O
13
Phenytoini
23
30
30 [4]
(22–39)
9.1 [5]
––j
>100
6.7
0.6
Phenobarbitali
Valproatei
61 [0.5]
(44–96)
(7.6–12)
490 [0.5]
(350–730)
280 [0.5]
(190–350)
a The compounds were administered orally.
b MES ¼ maximal electroshock seizure test in mg/kg. Numbers in parenthesis are 95% confidence intervals. The dose effect data was obtained at the ‘time of peak effect’ (indicated in hours in the brackets).
c Tox ¼ neurologic toxicity determined from the observation of motor impairment in mg/kg. Numbers in parenthesis are 95% confidence intervals. The dose effect data was obtained at the ‘time of peak
effect’ (indicated in hours in the brackets).
d PI ¼ protective index (TD50 /ED50).
e Ref. 1b.
f Ref. 1j.
g Ref. 2a.
h Not prepared.
i Ref. 17.
j No ataxia observed up to 3000 mg/kg.

3086
C. Bꢀeguin et al. / Bioorg. Med. Chem. 12 (2004) 3079–3096
most C(2) phenyl AAA (15, 16, 18, 19, 22) displayed
excellent activity in rats (ED50 < 30 mg/kg), and the most
potent compound, 19 (ED50 ¼ 7.1 mg/kg), was seven
times more potent than the corresponding FAA, 32
(ED50 ¼ 48 mg/kg).1b The anticonvulsant activity of 19
exceeded that of phenytoin (ED50 ¼ 30 mg/kg) and
phenobarbital (ED50 ¼ 9.1 mg/kg).16;17 The remaining
analogues (14, 20, 21, and 23), excluding 17, showed
moderate activity (ED50 ¼ 30–100 mg/kg). Finally, in the
C(2) methoxymethyl series (24–30) a narrow range of
activity was observed in rats (ED50 ¼ 22–35 mg/kg),
except for the tertiary amine 30 in which a slightly lower
anticonvulsant activity was found (ED50 ¼ 30–100 mg/
kg). The most active AAA in this subset were 26 and 27
(ED50 ¼ 22 mg/kg), which were nearly six times less
potent than 2.1j The neurological toxicities for AAA in
rats were generally low. We found increased toxicities
for the C(2) phenyl series (14–23) compared with the
C(2) methyl (4–13) and C(2) methoxymethyl (24–30)
analogues. However, our most active compound, 19,
provided a high protective index (PI ¼ 33) that com-
pared favorably with both phenytoin and phenobarbi-
tal.16;17
in mice (ip) that (R,S)-26 (ED50 ¼ 51mg/kg) was
more effective in controlling seizures than (R)-26
(ED50 > 100 mg/kg) while both compounds displayed
comparable activities in rats (po) [ED50 values (mg/kg):
(R,S)-26, 22 mg/kg; (R)-26, ꢀ30mg/kg]. At this stage, we
do not have sufficient data to provide a rationale for this
finding since we have not prepared and tested (S)-26. We
do note that both C(2) substituents (i.e., CH2OCH3,
N(H)CH2CH3) in 26 are of comparable size, possibly
permitting both enantiomers to bind to the putative
receptor.
The mechanism of action of FAA has remained elusive.
Radioligand displacement assays10;18 have failed to
identify receptor sites that are affected by FAA binding.
Thus, whole animal pharmacological profiles have been
used to distinguish these novel agents from established
antiepileptic agents. Previous findings have shown that
the iv infusion Metrazol test in mice was useful to
identify compounds with proconvulsant potential. We
tested our most active AAA, 19, further in this model to
ascertain if 19 and 32 share a common mechanism (or
mechanisms) of action. Compound 32 did not affect the
seizure threshold for minimal seizures induced by iv
infusion of Metrazol,7 while the seizure threshold for 19
was lowered with increasing doses (36 and 72 mg/kg,
data not shown).19 Therefore, 19 exhibited proconvul-
sant activity in the iv infusion Metrazol model while at
the same time preventing generalized tonic seizures. This
profile may appear to be unusual but has been previ-
ously observed for several other compounds including a
few marketed drugs (e.g., Mexiletine, Lidocaine), and
certain 5-HT blocking agents.20
Little differences in AAA activity were observed among
the three classes of compounds [C(2) methyl, C(2) phenyl,
C(2) methoxymethyl], and the observed potencies of the
compounds were not measurably affected by the degree
and type of terminal amine substitution. For their FAA
counterparts we observed distinctive SAR patterns.1
Why the absence of similar trends for AAA? We suggest
that several factors contributed to the observed activity
profile of AAA in rodents. AAA are basic compounds
and the N-terminal substituents are prone to metabolism.
By comparison, FAA are neutral and (R)-2 was shown to
undergo little metabolic change either in rodents or dogs.
In phase I human trials (R)-2 had nearly 100% oral bio-
availability.18 Thus, AAA and FAA are likely to have
different pharmacokinetic and metabolic properties that
affect drug concentrations in the brain. In Section 2.4, we
test this hypothesis, in part, by monitoring the in vitro
metabolism of a series of C(2) phenyl AAA.
2.4. Metabolism studies
We hypothesized that the AAA N-terminal substituent
may undergo metabolism explaining, in part, the lack of
distinctive SAR trends for this class of compounds.
Thus, we conducted a preliminary investigation using an
hepatic microsomal assay to determine the ease of sub-
strate consumption and the identity of key metabolic
products. We selected AAA candidates from the C(2)
phenyl series since they exhibited appreciable UV
absorbance amenable to LC–UV detection (k ¼ 256 nm).
A representative subset of compounds was chosen that
contained primary (14), secondary (15), and tertiary (19,
22) amino substituents. We included 46 in our study, the
hexadeuterio-analogue of 19, to aid our understanding
of the main metabolic process for this tertiary AAA.
Optically pure AAA 4, 24, and 26 were evaluated using the
MES test and their ED50 values were compared with the
corresponding racemic mixtures (Table 3). For 4 in mice
(ip), we observed an increase in anticonvulsant activity
proceeding from (S)-4 (ED50 > 300 mg/kg) to (R,S)-4
(ED50 ¼ 100–300 mg/kg) to (R)-4 (ED50 ¼ 10–30 mg/kg).
This trend is similar to that previously reported for FAA
31.1d The preference for (R) versus (S) stereochemistry was
also observed in rats (po) where (R)-4 (ED50 ¼ 19mg/kg)
was more active than (S)-4 (ED50 > 80mg/kg). We found
that (R)-4 and (R,S)-4 (ED50 ¼ 14mg/kg) displayed nearly
identical seizure protection. The stereochemical preference
for the (R)-isomer was also found for 24 where (R)-24
showed enhanced activity compared with (R,S)-24 in both
mice and rats. The ED50 values in mice for (R)-24 and
(R,S)-24 were 48 and 84mg/kg, respectively, while in rats
the ED50 values for (R)-24 and (R,S)-24 were 18 and 29mg/
kg, respectively. Surprising to us was the finding that the
N-ethyl analogue of 24, compound 26, did not follow the
pattern observed for 4, 24, and FAA.1c–e;g;j;l Here, we found
The C(2) phenyl AAA and their corresponding FAA 32
were incubated with either mice or rat liver microsomes
containing cytochrome P-450 enzymes in the absence
and presence of NADPH (22 ꢁC, 0–60 min), and sub-
strate disappearance was monitored by HPLC. Metab-
olites were identified by LC–UV and LC–MS/MS. Mass
detection was accomplished by electrospray ionization
(ESI).
Optimization of the HPLC gradient conditions (15 min
run) allowed the separation of most peaks. Molecular

C. Bꢀeguin et al. / Bioorg. Med. Chem. 12 (2004) 3079–3096
Table 3. Effect of stereochemistry at the C(2) position on AAA anticonvulsant activity
3087
Compd
C(2)
Mice (ip)a
Rat (po)b
MES,c ED50
>100, <300
Tox,d TD50
PIe
––
MES,c ED50
Tox,d TD50
PIe
CH3
H
N
Ph
(R,S)-4
>300
14 [1]
>500
>36
H2N
O
(7–22)
CH3
H
N
Ph
Ph
(R)-4
>10, <30
>300
>100, <300
––
––
19 [2]
(13–25)
>30
>80
>1.5
H2N
H2N
O
CH3
H
N
(S)-4
>300
>80
––
O
OCH3
H
(R,S)-24
84 [0.25]
(65–97)
290 [0.25]
(240–320)
3.5
29 [1]
(18–45)
>450
>16
N
Ph
Ph
H2N
H2N
O
OCH3
(R)-24
48 [0.25]
(40–61)
>30, <100
––
18 [4]f
>500
>28
H
N
O
OCH3
H
(R,S)-26
(R)-26
51 [0.25]
(42–61)
150 [0.25]
(120–180)
2.9
22 [0.25]
(13–33)
>500
>30
>23
––
N
Ph
N
H
O
OCH3
H
>100
>100, <300
––
ꢀ30 [0.5]
N
Ph
N
H
O
a The compounds were administered intraperitoneally.
b The compounds were administered orally.
c MES ¼ maximal electroshock seizure test in mg/kg. Numbers in parenthesis are 95% confidence intervals. The dose effect data was obtained at the
‘time of peak effect’ (indicated in hours in the brackets).
d Tox ¼ neurologic toxicity determined from rotorod test in mice and the observation of motor impairment in rats in mg/kg. Numbers in parenthesis
are 95% confidence intervals. The dose effect data was obtained at the ‘time of peak effect’ (indicated in hours in the brackets).
e PI ¼ protective index (TD50/ED50).
f Ref. 6.
ions in the LC–MS profile were obtained for the sub-
strates and most metabolites. We characterized most
metabolites formed after 60 min microsomal incubation.
Calibration curves were obtained using UV detection for
the AAA and 32 and were linear over a 5–500 lM range,
thus permitting us to quantify the amount of substrate
depletion over time. We recognized that without a
complete set of metabolite calibration curves we would
not be able to provide precise percentages of each
metabolite produced. Thus, our assignment of the major
and minor metabolic adducts rested upon the assump-
tion that these compounds displayed comparable UV
and ionization properties, and the near identical data set
obtained with UV and MS detection.
looked at differences in mass between isotope and non-
isotope AAA by using D2O solutions to help explain the
fragmentation patterns. Most AAA metabolites were
identified by their MS/MS spectra. The structures of
these metabolites and their HPLC retention time are
given in Table 4. We observed three metabolic out-
comes: N-dealkylation (type 1), N-benzylamide
hydroxylation (type 2), and N-oxide formation (type 3).
N-Dealkylation (type 1) metabolites 14, 15, and 16 were
readily identified by coinjection of authentic samples in
the HPLC with the reaction samples, and their struc-
tures were confirmed by mass. Metabolite 52 was iden-
tified by its mass and MS/MS fragmentation pattern.
Type 2 metabolites (Table 4) obtained from microsomal
incubations of 14, 15, 19, 22, 32, and 46 had early
retention times (min: 47: 5.3; 48: 3.2; 49: 3.4; 50: 4.6; 53:
4.3; 55: 4.9). The MS/MS fragmentation patterns for
Characteristic fragmentation patterns in the mass spec-
tra were observed for 14, 15, 19, 22, 32, and 46. Our
compounds contained exchangeable protons. Thus, we

Table 4. Proposed metabolism profile for N-substituted C(2) phenyl amino acid N0-benzylamidesa
Substrate
Compd
Substrate depletionb, %
N-Dealkylation metabolites (type 1)
Minor metabolites (type 2)
Minor metabolites (type 3)
––
32
ꢀ10
––
47 (5.3)
(12.0)
Ph
H
Ph
H
N
14
15
19
ꢀ25
ꢀ32
ꢀ75
––
––
N
Ph
OH
Ph
Ph
Ph
H2N
H2N
O
O
48 (3.2)
(6.5)
Ph
Ph
Ph
H
H
H
N
Ph
OH
N
Ph
N
––
N
H
H2N
N
H
O
O
O
(7.1)
Ph
49 (3.4)
14 (6.6)
Ph
Ph
Ph
Ph
O
N
H
H
N
H
H
N
H
N
N
Ph
OH
Ph
Ph
Ph
N
N
N
H
H2N
N
O
O
O
O
O
15 (7.1)
14 (6.6)
50 (4.6)
51 (7.3)
(11.0)
Ph
O
Ph
Ph
H
Ph
H
Ph
N
H
H
N
H
N
CD3
CD3
N
Ph
N
Ph
N
Ph
CD3
Ph
CD3
Ph
OH
46
––c
N
H
H2N
N
N
O
O
O
O
CD3
O
CD3
CD3
52 (7.0)
14 (6.6)
53 (4.3)
(10.2)
54 (7.3)
Ph
Ph
H
H
22d
ꢀ52
––
––
N
Ph
OH
N
Ph
N
N
O
O
55 (4.9)
(11.2)
a The number in parenthesis corresponds to the HPLC retention time in min.
b Substrate depletion determined by UV after 60 min. The percentage is the average between the values obtained after mice and rats microsomal incubations.
c Compound 46 showed a comparable depletion of substrate when tested against 19.
d Other metabolites (56, 57) were identified for 22 and are not included in this chart.

C. Bꢀeguin et al. / Bioorg. Med. Chem. 12 (2004) 3079–3096
3089
those metabolites were characterized by the loss of 106
and 151 (data not shown). The 151 Da unit corre-
sponded to the elimination of an ‘oxygenated’ 135 Da
fragment previously observed for substrates 19, 22, and
46 (data not shown). This finding suggested the presence
of an OH group within the N-benzylamide unit. Further
supporting this contention was the detection of a second
fragment ion that equated to the loss of 106 from the
molecular ion. This fragmentation pattern suggests that
hydroxylation occurred either at the methylene position
of the benzylamide unit or within the aromatic unit. The
early retention times in the HPLC are consistent with
either metabolite and there is prior literature precedence
for both.21 Significantly, all of the type 2 metabolic
processes observed for our C(2) phenyl AAA corre-
sponded to N-benzylamide hydroxylation. Under our
assay conditions, we did not detect C(2) phenyl
hydroxylation (phenolic)-derived products.
ylation to give 56. The two peaks in the HPLC (10.6,
10.9 min) may correspond to a pair of diastereoisomers
where hydroxylation occurred at one of the pyrrolidine
ring carbons, or the two peaks may correlate to two
metabolic products where hydroxylation occurred at
different positions within the pyrrolidine ring. Further-
more, enzymatic oxidation proceeded to give metabolite
57 (HPLC: 13.6 min). The MS/MS data did not allow us
to identify the position of the hydroxyl or the carbonyl
unit in the oxidized pyrrolidine ring. These observations
were in general agreement with those previously
reported for the metabolism of nicotine (58).24 For 58,
hydroxylation proceeded selectively at the methylene
group adjacent to the N-methyl position.
Type 3 metabolites were observed for only 19 and 46.
The fragmentation patterns for type 3 metabolites were
consistent with the formation of N-oxides 51 and 54.
The major MS/MS fragments were the loss of 60 for
metabolite 51 and 66 for the deuteriated adduct 54 to
provide m=z 225 and suggested the molecular cleavage of
the N-oxide units. Significantly, the detection of frag-
mentation ions at m=z ¼ 107 for these type 3 metabolites
indicated the presence of an unmodified N-benzylamine
unit. The LC–MS/MS analysis of a synthetic sample of
N-oxide 51 confirmed the identity of type 3 metabolite
(data not shown).
When we monitored the rate of substrate depletion we
found little differences for the mice (Fig. 3A) and the
rats (Fig. 3B). Accordingly, we suspect that the agent(s)
responsible for the control of MES-induced seizures is
the same in both mice (ip) and rats (po). In agreement
with this notion, we observed the same order of anti-
convulsant activity (19 > 15, 22 > 14) in both animals
(Tables 1 and 2).
Replacement of the dimethylamino unit in 19 by the
fully deuteriated dimethylamino group to give 46 neither
affected the product profile nor the in vitro metabolism
rate (data not shown). This finding suggested that N-
demethylation occurred by a rate-determining one
electron transfer mechanism, a pathway previously
suggested by Macdonald and co-workers.25
We observed that the in vitro metabolism profiles for 14,
15, 19, 22, 32, and 46 were similar for both mice and rats
microsomes. Incubation of 14, 15, 19, 22, and 32 with
the microsomes in the absence of NADPH gave very
little or no consumption of substrate (<2%) and sug-
gested little nonoxidative metabolism for this class of
compounds. Addition of NADPH (1 mM) to the incu-
bation medium led to increased levels of metabolism and
the extent of reaction was substrate dependent. FAA 32
showed little change (ꢀ10%) over the 60-min reaction
time course. The principal metabolite for 32 resulted
from N-benzylamide hydroxylation. Our finding that 32
underwent little in vitro metabolism was in agreement
with previous results obtained for FAA (R)-2.18
Removal of the acetyl unit in 32 to give AAA 14 led to a
threshold increase in metabolism (60 min: est. 25%
substrate depletion) with N-benzylamide hydroxylation
accounting for the majority of the product. Successive
incorporation of methyl units at the AAA N-terminal
amine site to provide 15 and 19 led to a steady, further
increase in the extent of metabolism (60 min: 15: est.
32%, 19: est. 75%). The major pathway for AAA 15, 19,
and 46 consumption was N-dealkylation. For 15, 19,
and 46 N-benzylamide hydroxylation (type 2) was
observed; while for 19 and 46, N-oxide formation (type
3) accounted for a substantial portion of the depleted
substrate. N-Oxide formation is a common metabolism
pathway for tertiary amines22;23 and we were surprised
that we did not detect any N-oxide product for AAA 22.
Inclusion of the N-terminal group within a cyclic pyr-
rolidine ring to give 22 led to pyrrolidine ring hydrox-
3. Conclusions
AAA displayed moderate-to-excellent activity in the
MES test and like their FAA counterparts were devoid
of activity in the scMet test.1 In select cases the anti-
convulsant activities exceeded that of their FAA coun-
terpart. Review of the pharmacological data showed
that seizure protection in rats (po) generally surpassed
the effect observed in mice (ip). The anticonvulsant
activity for 19 in rats (po) exceeded that of phenytoin
and phenobarbital17;18 and provided a PI ratio of 33.
Our results demonstrated that AAA 14, 16, 19, 22, and
46 have a more complex in vitro metabolic profile than
the corresponding FAA 32. We show that AAA
undergo extensive metabolic change where modification
of the N-terminal substituent constitutes the major
process. Accordingly, we suspect that the observed
whole animal pharmacological data for the individual
AAA may correspond to a composite set of activities
that includes the AAA and the corresponding AAA
metabolites and which is further modulated by
the ADMET properties of these compounds. For

3090
C. Bꢀeguin et al. / Bioorg. Med. Chem. 12 (2004) 3079–3096
Figure 3. In vitro metabolism rate of substrate consumption of C(2) phenyl AAA. Incubations were performed with 1 mg microsomal protein, 1 mM
NADPH, and 100 lM substrate at room temperature for 60 min.
C(2)-phenyl AAA both the anticonvulsant activities and
the extent of metabolism increased in proceeding from
primary (14) to secondary (15) to tertiary (19, 32)
amines. These trends diminished the beneficial effects of
N-alkyl substitution on biological activity and contrib-
uted, in part, to the absence of a distinctive SAR trend
for this class of compounds.
cipitate that formed during the reaction was filtered and
the filtrate was evaporated and purified by column
chromatography (SiO2; MeOH–CHCl3 or EtOAc–hex-
anes) to obtain the desired product.
4.3. Synthesis of N-benzyl-substituted amino acid amide.
Method B
A THF (45–125 mL) solution of carboxylic acid
(1 equiv) was cooled ()78 ꢁC) and then 4-meth-
ylmorpholine (1.25 equiv) was added and the solution
was stirred (2 min). Isobutyl chloroformate (1.25 equiv)
was then added and the reaction was stirred (2 min)
followed by the addition of benzylamine (1.25 equiv).
The reaction was stirred at )78 ꢁC (15 min), allowed to
warm to room temperature and then stirred (1.5 h). The
reaction mixture was filtered and the filtrate evaporated.
The residue was purified by column chromatography
(SiO2; 1:49, MeOH–CHCl3) to obtain the desired
product.
4. Experimental
4.1. General methods
Melting points were determined with a Thomas–Hoover
melting point apparatus and are uncorrected. Infrared
spectra (IR) were run on a ATI Mattson Genesis Series
FTIRTM spectrometer. Absorption values are expressed
in wave-numbers (cmÀ1). Optical rotations were
obtained on a Jasco P-1030 polarimeter. Proton (1H
NMR) and carbon (13C NMR) nuclear magnetic reso-
nance spectra were taken either on General Electric QE-
300 NMR or Varian Gemini 2000 spectrometers. Low
resolution mass spectra (CI+) were obtained with a
Varian MAT CH-5 spectrometer by Dr. M. Moini at the
University of Texas––Austin. The high-resolution
chemical ionization mass spectrum was performed on a
Finnigan MAT TSQ-70 by Dr. M. Moini at the Uni-
versity of Texas––Austin. Microanalysis were provided
by Atlantic Microlab, Inc. (Norcross, GA).
4.4. Synthesis of( R,S)-N-benzyl-2-bromo-3-methoxypro-
pionamide (45)
Compound 45 (1.17 g, 73%) was prepared as a white
solid from 4411 (1.06 g, 5.89 mmol), 4-methylmorpholine
(810 lL, 7.37 mmol), isobutyl chloroformate (960 lL,
7.37 mmol), and benzylamine (805 lL, 7.37 mmol) uti-
lizing method B: mp 71–73 ꢁC; Rf 0.72 (1:49, MeOH–
1
4.2. Synthesis of N0-alkyl-substituted N-benzyl amino acid
amides. Method A
CHCl3); H NMR (CDCl3) d 3.43 (s, OCH3), 3.87 (dd,
J ¼ 5:0, 10.7 Hz, CHH0O), 3.95 (dd, J ¼ 4:6, 10.7 Hz,
CHH0O), 4.43–4.50 (m, CH and CH2Ph), 6.93 (br s,
NH), 7.24–7.37 (m, 5PhH); 13C NMR (CDCl3) 44.0,
47.9, 59.2, 73.6, 127.4, 127.5, 128.7, 137.4, 167.0 ppm;
MS (+CI) (rel intensity) 275 (12), 274 (90), 273 (14), 272
A
solution of N-benzyl-2-haloamino acid amide
(1 equiv) in amine (excess) was stirred in a Schlenck
flask. In select cases, THF was added. The white pre-

C. Bꢀeguin et al. / Bioorg. Med. Chem. 12 (2004) 3079–3096
3091
(Mþ+1, 100), 102 (17); Mr (+CI) 272.02884 [Mþ+1]
(calcd for C11H1579BrNO2 272.02861). Anal. Calcd for
C11N14BrNO2: C, 48.55; H, 5.18; N, 5.15. Found: C,
48.68; H, 5.14; N, 5.13.
7.36 (m, 5PhH); 13C NMR (CDCl3) 14.2, 40.9, 42.9,
59.7, 67.3, 127.3, 127.6, 128.5, 138.3, 172.8 ppm; MS
(+CI) (rel intensity) 224 (14), 223 (Mþ+1, 100); Mr (+CI)
223.14492 [Mþ+1] (calcd for C12H19N2O2 223.14465).
Anal. Calcd for C12H18N2O2: C, 64.84; H, 8.16; N,
12.60. Found: C, 64.51; H, 8.14; N, 12.20.
4.5. Synthesis of( R,S)-N-benzyl-2-(N-methoxylamino)-
propionamide (7)
4.8. Synthesis of( R,S)-N-benzyl-2-[bis(2-methoxy-
ethyl)amino]propionamide (11)
Compound 7 (975 mg, 64%) was prepared as a pale-
yellow oil from 3826 (1.77 g, 7.31 mmol), O-meth-
ylhydroxylamine (4.0 mL, 76.5 mmol), and THF (5 mL)
at 45 ꢁC (4 d) utilizing method A. Purification was by
column chromatography (SiO2; 1:49, MeOH–CHCl3):
Rf 0.36 (1:49, MeOH–CHCl3); IR (neat) 3291 (br), 3087,
A solution of 3826 (1.93 g, 7.98 mmol) in bis(2-meth-
oxyethyl)amine (12 mL, 81.27 mmol) was stirred at 65 ꢁC
(4 h). EtOAc (100 mL) was added and then the reaction
mixture was extracted with an aq 0.1 M KH2PO4 solu-
tion (5 · 20 mL). The organic layer was dried (Na2SO4)
and evaporated. The residue was purified by column
chromatography (SiO2; 1:49, MeOH–CHCl3) to obtain
1.03 g (44%) of pure 11 as a pale-yellow oil: Rf 0.30
(1:49, MeOH–CHCl3); IR (neat) 3319 (br), 1665,
3031, 1655, 1537 cmÀ1 1H NMR (CDCl3) d 1.27 (d,
;
J ¼ 7:2 Hz, CH3), 3.49 (s, OCH3), 3.63 (dq, J ¼ 5:4,
7.2 Hz, CH), 4.44 (dd, J ¼ 5:7, 15.0 Hz, CHH0Ph), 4.54
(dd, J ¼ 6.3, 15.0 Hz, CHH0Ph), 5.64 (d, J ¼ 5:4 Hz,
NH), 6.98 (br s, NH), 7.25–7.38 (m, 5PhH); 13C NMR
(CDCl3) 15.7, 43.2, 60.2, 62.0, 127.4, 127.6, 128.7, 138.2,
173.3 ppm; MS (+CI) (rel intensity) 210 (14), 209
(Mþ+1, 100); Mr (+CI) 209.13004 [Mþ+1] (calcd for
C11H17N2O2 209.12900). Anal. Calcd for C11H16-
N2O2Æ0.1H2O: C, 62.90; H, 7.77; N, 13.34. Found: C,
62.75; H, 7.69; N, 13.20.
1517 cmÀ1
;
1H NMR (CDCl3) d 1.28 (d, J ¼ 6:9 Hz,
CH3), 2.61–2.71 (m, 2CH2CH2OCH3), 3.20 (s, 2OCH3),
3.30–3.44 (m, 2CH2OCH3), 3.50 (q, J ¼ 6:9 Hz, CH),
4.30–4.52 (m, CH2Ph), 7.22–7.35 (m, 5PhH), 8.13 (br s,
NH); 13C NMR (CDCl3) 7.8, 43.3, 50.8, 58.7, 60.7, 71.2,
127.0, 127.8, 128.9, 138.9, 174.1 ppm; MS (+CI) (rel
intensity) 296 (15), 295 (Mþ+1, 100); Mr (+CI)
295.20147 [Mþ+1] (calcd for C16H27N2O3 295.20217).
Anal. Calcd for C16H26N2O3Æ0.3H2O: C, 64.10; H. 8.94;
N, 9.34. Found: C, 64.13; H, 8.89; N, 9.15.
4.6. Synthesis of( R,S)-N-benzyl-2-(2-methoxyethyl-
amino)propionamide (8)
Compound 8 (1.25 g, 95%) was prepared as an oil from
3710 (1.10 g, 5.58 mmol) and 2-methoxyethylamine
(20 mL, 230 mmol) at 50 ꢁC (2 d) utilizing method A.
Purification was by column chromatography (SiO2;
1:24, MeOH–CHCl3): Rf 0.39 (1:24, MeOH–CHCl3); IR
4.9. Synthesis of( R,S)-N-benzyl-2-pyrrolidin-1-yl-propion-
amide (12)
Compound 12 (1.12 g, 86%) was prepared as an oil from
3710 (1.10 g, 5.58 mmol) and pyrrolidine (20 mL,
240 mmol) at room temperature (18 h) utilizing method
A. Purification was by column chromatography (SiO2;
1:24, MeOH–CHCl3): Rf 0.10 (1:2, EtOAc–hexanes); IR
1
(neat) 3313 (br), 3064, 3031, 1656, 1525 cmÀ1; H NMR
(CDCl3) d 1.35 (d, J ¼ 6:9 Hz, CH3), 2.18 (br s, NH),
2.63–2.71
(m,
NHCHH0CH2),
2.80–2.88
(m,
NHCHH0CH2), 3.28 (q, J ¼ 6:9 Hz, CH), 3.29 (s,
OCH3), 3.35–3.48 (m, CH2OCH3), 4.44 (d, J ¼ 6:0 Hz,
CH2Ph), 7.25–7.32 (m, 5PhH), 7.77 (t, J ¼ 6:0 Hz,
NHCH2Ph); 13C NMR (CDCl3) 19.6, 42.9, 47.8, 58.1,
58.6, 71.4, 127.2, 127.5, 128.5, 138.5, 174.6 ppm; MS
(+CI) (rel intensity) 238 (15), 237 (Mþ+1, 100); Mr (+CI)
237.16040 [Mþ+1] (calcd for C13H21N2O2 237.16030).
Anal. Calcd for C13H20N2O2Æ0.31H2O: C, 64.55; H, 8.59;
N, 11.58. Found: C, 64.87; H, 8.55; N, 11.72.
(neat) 3300 (br), 2969, 1663, 1518 cmÀ1 1H NMR
;
(CDCl3) d 1.33 (d, J ¼ 6:9 Hz, CH3), 1.71–1.75 (m,
CH2CH2), 2.51–2.58 (m, CH2NCH2), 2.97 (q,
J ¼ 6:9 Hz, CH), 4.43–4.48 (m, CH2Ph), 7.24–7.33 (m,
5PhH), the signal for the NH proton was not detected;
13C NMR (CDCl3) 17.1, 23.3, 42.8, 51.2, 63.9, 127.2,
127.5, 128.5, 138.6, 174.7 ppm; MS (+CI) (rel intensity)
234 ppm (15), 233 (Mþ+1, 100); Mr (+CI) 233.16523
[Mþ+1] (calcd for C14H21N2O 233.16539). Anal. Calcd
for C14H20N2OÆ0.38H2O: C, 70.31; H, 8.75; N, 11.71.
Found: C, 70.26; H, 8.51; N, 11.77.
4.7. Synthesis of( R,S)-N-benzyl-2-(N-methoxy-N-meth-
ylamino)propionamide (10)
4.10. Synthesis of( R,S)-N-benzyl-2-morpholin-4-yl-prop-
ionamide (13)
Compound 10 (1.18 g, 89%) was prepared as a semi-
solid from 3826 (1.44 g, 5.95 mmol), N,O-dimethylhydr-
oxylamine (5 mL, 84.59 mmol), and THF (10 mL) at
45 ꢁC (4 d) utilizing method A. Purification was by col-
umn chromatography (SiO2; 1:49, MeOH–CHCl3): Rf
0.50 (1:49, MeOH–CHCl3); IR (neat) 3311 (br), 1661,
Compound 13 (1.25 g, 95%) was prepared as an off-white
solid from 3710 (1.10 g, 5.58 mmol) and morpholine
(20 mL, 229 mmol) at 45 ꢁC (18 h) utilizing method A.
Purification was by column chromatography (SiO2; 1:24,
MeOH–CHCl3): mp 58–59 ꢁC; Rf 0.80 (1:24, MeOH–
1525 cmÀ1
;
1H NMR (CDCl3) d 1.32 (d, J ¼ 6:6 Hz,
1
CHCl3); IR (KBr) 3276 (br), 3068, 1648, 1548 cmÀ1; H
CH3), 2.58 (s, NCH3), 3.32 (q, J ¼ 6:6 Hz, CH), 3.43 (s,
OCH3), 4.41–4.54 (m, CH2Ph), 6.98 (br s, NH), 7.24–
NMR (CDCl3) d 1.27 (d, J ¼ 6:9 Hz, CH3), 2.42–2.57

3092
C. Bꢀeguin et al. / Bioorg. Med. Chem. 12 (2004) 3079–3096
(m, CH2NCH2), 3.60–3.72 (m, CH2OCH2), 3.06 (q,
J ¼ 6:9 Hz, CH), 4.39–4.52 (m, CH2Ph), 7.24–7.35 (m,
5PhH), 7.48 (br s, NH); 13C NMR (CDCl3) 12.1, 42.9,
50.2, 64.4, 66.9, 127.3, 127.4, 128.6, 138.5, 173.3 ppm;
MS (+CI) (rel intensity) 250 (15), 249 (Mþ+1, 100); Mr
(+CI) 249.16088 [Mþ+1] (calcd for C14H21N2O2
249.16030). Anal. Calcd for C14H20N2O2: C, 67.72; H,
8.12; N, 11.28. Found: C, 67.48; H, 8.18; N, 11.36.
(m, NCHH0), 3.28 (s, OCH3), 3.43–3.46 (m, CH2OCH3),
4.24 (s, CH), 4.42 (dd, J ¼ 4:9, 13.7 Hz, CHH0Ph), 4.49
(dd, J ¼ 4:4, 13.7 Hz, CHH0Ph), 7.21–7.41 (m, 10PhH),
7.65 (br s, NH), the signal for the remaining NH proton
was not detected; 13C NMR (CDCl3) 43.1, 48.1, 58.7,
67.7, 71.5, 127.2, 127.3, 127.6, 128.0, 128.6, 128.7, 138.4,
139.4, 172.1 ppm; MS (+CI) (rel intensity) 300 (20), 299
(Mþ+1, 100), 164 (11); Mr (+CI) 299.17653 [Mþ+1] (calcd
for C18H23N2O2 299.17595). Anal. Calcd for
C18H22N2O2Æ0.1H2O: C, 72.02; H, 7.45; N, 9.33. Found:
C, 71.86; H, 7.47; N, 9.39.
4.11. Synthesis of( R,S)-N-benzyl-2-ethylamino-2-phenyl-
acetamide (16)
Compound 16 (1.60 g, 78%) was prepared as a thick oil
from 39 (2.00 g, 7.71 mmol) and ethylamine (2.0 M in
THF, 25 mL, 50 mmol) at 45 ꢁC (5 d) utilizing method A.
Purification was by column chromatography (SiO2;
1:49, MeOH–CHCl3): Rf 0.18 (1:49, MeOH–CHCl3); IR
4.14. Synthesis of( R,S)-N-benzyl-2-dimethylamino-2-
phenylacetamide (19)
Compound 19 (3.07 g, 90%) was prepared as a white
solid from 39 (3.30 g, 12.71 mmol) and dimethylamine
(2.0 M in THF, 30 mL, 60 mmol) at 40 ꢁC (18 h) utilizing
method A. Purification was by column chromatography
(SiO2; 1:49, MeOH–CHCl3): mp 88–89 ꢁC; Rf 0.33 (1:49,
MeOH–CHCl3); IR (KBr) 3228 (br), 3048, 1645,
1
(neat) 3296 (br), 3063, 3034, 1661, 1521 cmÀ1; H NMR
(CDCl3) d 1.17–1.25 (m, CH2CH3), 2.70–2.84 (m,
CH2CH3), 4.34 (s, CH), 4.58 (d, J ¼ 5:7 Hz, CH2Ph),
7.33–7.53 (m, 10PhH), 7.73 (br s, NH), the signal for the
remaining NH proton was not detected; 13C NMR
(CDCl3) 15.3, 43.0, 43.2, 67.8, 127.1, 127.3, 127.5, 128.0,
128.6, 128.7, 138.4, 138.7, 172.3 ppm; MS (+CI) (rel
intensity) 270 (16), 269 (Mþ+1, 100); Mr (+CI)
269.16475 [Mþ+1] (calcd for C17H21N2O 269.16539).
Anal. Calcd for C17H20N2OÆ0.2H2O: C, 75.80; H, 7.56;
N, 10.30. Found: C, 74.96; H, 7.28; N, 10.24.
1
1543 cmÀ1; H NMR (CDCl3) d 2.18 (s, N(CH3)2), 3.77
(s, CH), 4.43 (dd, J ¼ 5:5, 14.7 Hz, CHH0Ph), 4.49 (dd,
J ¼ 6:3, 14.7 Hz, CHH0Ph), 7.22–7.35 (m, 10PhH), the
signal for the NH proton was not detected; 13C NMR
(CDCl3) 43.2, 43.8, 77.1, 127.4, 127.8, 128.0, 128.4,
128.6, 128.7, 136.6, 138.4, 171.6 ppm; MS (+CI) (rel
intensity) 270 (18), 269 (Mþ+1, 100), 134 (20); Mr (+CI)
269.16625 [Mþ+1] (calcd for C17H21N2O 269.16539).
Anal. Calcd for C17H20N2O: C, 76.09; H, 7.51; N, 10.44.
Found: C, 76.15; H, 7.48; N, 10.48.
4.12. Synthesis of( R,S)-N-benzyl-2-(N-methoxylamino)-
2-phenylacetamide (17)
Compound 17 (327 mg, 35%) was prepared as a pale-
yellow oil from 41 (1.10 g, 3.45 mmol) and O-methyl-
hydroxylamine (4.0 mL, 76.5 mmol) at 50 ꢁC (2 d) utiliz-
ing method A. Purification was by column chromatog-
raphy (SiO2; 1:1, EtOAc–hexanes): Rf 0.52 (1:1, EtOAc–
hexanes); IR (neat) 3296 (br), 3066, 1662, 1534 cmÀ1; 1H
NMR (CDCl3) d 3.51 (s, OCH3), 4.46 (dd, J ¼ 5:5,
15.0 Hz, CHH0Ph), 4.56 (dd, J ¼ 6:0, 15.0 Hz, CHH0Ph),
4.62 (s, CH), 6.85 (br s, NH), 7.23–7.40 (m, 10PhH), the
signal for the remaining NH proton was not detected; 13C
NMR (CDCl3) 43.2, 61.9, 69.1, 127.4, 127.5, 127.8, 128.6,
128.8, 135.4, 138.0, 170.5 ppm, one aromatic carbon
signal was not detected and is believed to overlap with
nearby peaks; MS (+CI) (rel intensity) 272 (17), 271
(Mþ+1, 100), 267 (14), 240 (18), 239 (94); Mr (+CI)
271.14480 [Mþ+1] (calcd for C16H19N2O2 271.14465).
Anal. Calcd for C16H18N2O2Æ0.3H2O: C, 69.70; H, 6.80;
N, 10.16. Found: C, 69.71; H, 6.78; N, 10.14.
4.15. Synthesis of( R,S)-N-benzyl-2-(N-methoxy-N-
methylamino)-2-phenylacetamide (20)
Compound 20 (1.34 g, 75%) was prepared as an off-
white solid from 41 (2.00 g, 6.27 mmol) and N,O-di-
methylhydroxylamine (7 mL, 118.43 mmol) at 60 ꢁC
(2 d) utilizing method A. Purification was by column
chromatography (SiO2; 1:49, MeOH–CHCl3): mp 108–
109 ꢁC; Rf 0.16 (1:4, EtOAc–hexanes); IR (KBr) 3302,
3079, 1659, 1555 cmÀ1 1H NMR (CDCl3) d 2.45 (s,
;
NCH3), 3.43 (s, OCH3), 4.17 (s, CH), 4.43 (dd, J ¼ 5:8,
14.9 Hz, CHH0Ph), 4.51 (dd, J ¼ 6:0, 14.9 Hz, CHH0Ph),
6.93 (br s, NH), 7.22–7.40 (m, 10PhH); 13C NMR
(CDCl3) 41.8, 43.2, 59.5, 78.4, 127.4, 127.7, 128.4, 128.6,
135.8, 138.3, 170.5 ppm, two aromatic carbon signals
were not detected and are believed to overlap with
nearby peaks; MS (+CI) (rel intensity) 286 (15), 285
(Mþ+1, 100), 253 (12), 150 (50), 120 (10); Mr (+CI)
285.16101 [Mþ+1] (calcd for C17H21N2O2 285.16030).
Anal. Calcd for C17H20N2O2: C, 71.87; H, 7.09; N, 9.85.
Found: C, 71.78; H, 7.14; N, 9.80.
4.13. Synthesis of( R,S)-N-benzyl-2-(2-methoxyethyl-
amino)-2-phenylacetamide (18)
Compound 18 (1.81 g, 98%) was prepared as a thick oil
from 39 (1.61 g, 6.20 mmol) and 2-methoxyethylamine
(20 mL, 230 mmol) at 40 ꢁC (2 d) utilizing method A.
Purification was by column chromatography (SiO2; 1:49,
MeOH–CHCl3): Rf 0.29 (1:49, MeOH–CHCl3); IR (neat)
4.16. Synthesis of( R,S)-N-benzyl-2-[bis(2-methoxy-
ethyl)amino]-2-phenylacetamide (21)
3307 (br), 3062, 3034, 1662, 1523 cmÀ1
;
1H
A solution of 39 (1.50 g, 5.78 mmol) in bis(2-methoxy-
ethyl)amine (12 mL, 81.27 mmol) was stirred at 45 ꢁC
NMR (CDCl3) d 2.69–2.77 (m, NCHH0), 2.81–2.89

C. Bꢀeguin et al. / Bioorg. Med. Chem. 12 (2004) 3079–3096
3093
(2 d). EtOAc (100 mL) was added and then the reaction
mixture was extracted with an aqueous 0.1 M KH2PO4
solution (5 · 20 mL). The organic layer was dried
(Na2SO4) and evaporated. The residue was purified by
column chromatography (SiO2; 1:49, MeOH–CHCl3) to
obtain 1.43 g (69%) of pure 21 as a pale-yellow oil: Rf
0.24 (1:49, MeOH–CHCl3); IR (neat) 3309 (br), 3033,
C12H18N2O2Æ0.25H2O: C, 63.55; H, 8.22; N, 12.35.
Found: C, 63.49; H, 8.27; N, 12.34.
4.19. Synthesis of( R,S)-N-benzyl-2-ethylamino-3-meth-
oxypropionamide (26)
1663, 1524 cmÀ1 1H NMR (CDCl3) d 2.51–2.59 (m,
;
Compound 26 (2.19 g, 88%) was prepared as a pale-
yellow oil from 45 (2.88 g, 10.59 mmol) and ethylamine
(2.0 M in THF, 25 mL, 50 mmol) at 45 ꢁC (3 d) utilizing
method A. Purification was by column chromatography
(SiO2; 1:49, MeOH–CHCl3): Rf 0.32 (1:49, MeOH–
CHCl3); IR (neat) 3318 (br), 3066, 3033, 1661,
2CHH0CH2OCH3), 2.72–2.81 (m, 2CHH0CH2OCH3),
3.19 (s, 2OCH3), 3.31–3.44 (m, 2CH2OCH3), 4.48 (dd,
J ¼ 5:2, 14.9 Hz, CHH0Ph), 4.54 (dd, J ¼ 6:3, 14.9 Hz,
CHH0Ph), 4.62 (s, CH), 7.25–7.35 (m, 10PhH), 8.52 (br
s, NH); 13C NMR (CDCl3) 43.5, 51.4, 58.7, 70.9, 71.1,
127.1, 127.6, 127.9, 128.2, 129.7, 136.1, 138.8,
172.2 ppm, one aromatic carbon signal was not
detected and is believed to overlap with nearby peaks;
MS (+CI) (rel intensity) 358 (22), 357 (Mþ+1, 100); Mr
(+CI) 357.21801 [Mþ+1] (calcd for C21H29N2O3
357.21782). Anal. Calcd for C21H28N2O3: C, 70.76; H,
7.92; N, 7.86. Found: C, 70.49; H, 7.94; N, 7.77.
1526 cmÀ1
;
1H NMR (CDCl3) d 1.07 (t, J ¼ 7:2 Hz,
CH2CH3), 1.73 (br s, NH), 2.55–2.71 (m, CH2CH3),
3.30–3.36 (m, CH), 3.35 (s, OCH3), 3.53–3.59 (m,
CHH0OCH3), 3.64–3.68 (m, CHH0OCH3), 4.46 (d,
J ¼ 6:3 Hz, CH2Ph), 7.25–7.36 (m, 5PhH), 7.82 (br s,
NH), addition of excess (R)-())-mandelic acid27 to a
CDCl3 solution containing 26 gave two signals for the
methoxy methyl protons (d 3.02, 3.07) and the N-ethyl
methyl protons (d 1.01, 1.05); 13C NMR (CDCl3) 15.3,
42.9, 43.2, 58.7, 62.6, 72.6, 127.2, 127.4, 128.5, 138.4,
172.0 ppm; MS (+CI) (rel intensity) 238 (16), 237
(Mþ+1, 100), 102 (12); Mr (+CI) 237.16052 [Mþ+1]
(calcd for C13H21N2O2 237.16030). Anal. Calcd for
C13H20N2O2Æ0.15H2O: C, 65.33; H, 8.56; N, 11.72.
Found: C, 65.34; H, 8.61; N, 11.73.
4.17. Synthesis of( R,S)-N-benzyl-2-pyrrolidin-1-yl-2-
phenylacetamide (22)
Compound 22 (2.47 g, 99%) was prepared as a white
solid from 39 (2.20 g, 8.48 mmol) and pyrrolidine
(20 mL, 240 mmol) at 40 ꢁC (18 h) utilizing method A.
Purification was by column chromatography (SiO2;
1:49, MeOH–CHCl3): mp 96–97 ꢁC; Rf 0.36 (SiO2; 1:49,
MeOH–CHCl3); IR (KBr) 3328, 2952, 1655, 1517 cmÀ1
;
1H NMR (CDCl3) d 1.71–1.75 (m, CH2CH2), 2.42–2.54
(m, CH2NCH2), 3.85 (s, CH), 4.42 (dd, J ¼ 5:6, 15.0 Hz,
CHH0Ph), 4.49 (dd, J ¼ 6:1, 15.0 Hz, CHH0Ph), 7.19–
7.41 (m, 10PhH), the signal for the NH proton was not
detected; 13C NMR (CDCl3) 23.3, 43.1, 52.6, 75.6, 127.3,
127.6, 127.9, 128.1, 128.4, 128.6, 138.2, 138.4,
171.8 ppm; MS (+CI) (rel intensity) 296 (19), 295
(Mþ+1, 100); Mr (+CI) 295.18071 [Mþ+1] (calcd for
C19H23N2O 295.18104). Anal. Calcd for C19H22N2O: C,
77.52; H, 7.53; N, 9.52. Found: C, 77.75; H, 7.66; N,
9.57.
4.20. Synthesis of( R,S)-N-benzyl-2-methoxyamino-3-
methoxypropionamide (27)
Compound 27 (2.20 g, 83%) was prepared as a pale-
yellow oil from 45 (2.99 g, 10.99 mmol), O-methylhydr-
oxylamine (4.8 mL, 91.9 mmol), and THF (5 mL) at
45 ꢁC (5 d) utilizing method A. Purification was by col-
umn chromatography (SiO2; 1:49, MeOH–CHCl3): Rf
0.43 (1:49, MeOH–CHCl3); IR (neat) 3314 (br), 3064,
1
1660, 1533 cmÀ1; H NMR (CDCl3) d 3.28 (s, OCH3),
3.44 (s, NOCH3), 3.51–3.62 (m, CH2OCH3), 3.71–3.76
(m, CH), 4.38 (dd, J ¼ 6:0, 15.0 Hz, CHH0Ph), 4.46 (dd,
J ¼ 5:8, 15.0 Hz, CHH0Ph), 6.24 (d, J ¼ 5:7 Hz, NH),
7.18–7.30 (m, 5PhH), 7.48 (br s, NH); 13C NMR
(CDCl3) 42.4, 58.3, 61.2, 63.6, 69.4, 126.7, 126.8, 128.0,
137.8, 169.9 ppm; MS (+CI) (rel intensity) 240 (14), 239
(Mþ+1, 100), 207 (72); Mr (+CI) 239.13993 [Mþ+1]
(calcd for C12H19N2O3 239.13957). Anal. Calcd for
C12H18N2O3: C, 60.49; H, 7.61; N, 11.76. Found: C,
60.30; H, 7.68; N, 11.71.
4.18. Synthesis of( R,S)-N-benzyl-2-methylamino-3-
methoxypropionamide (25)
Compound 25 (1.42 g, 87%) was prepared as an oil from
45 (2.01 g, 7.39 mmol) and methylamine (2.0 M in THF,
25 mL, 50 mmol) at 45 ꢁC (2 d) utilizing method A.
Purification was by column chromatography (SiO2;
1:49, MeOH–CHCl3): Rf 0.31 (1:49, MeOH–CHCl3); IR
(neat) 3018 (br), 3064, 1656, 1525 cmÀ1 1H NMR
;
4.21. Synthesis of( R,S)-N-benzyl-2-(2-methoxyethyl-
amino)-3-methoxypropionamide (28)
(CDCl3) d 1.84 (br s, NH), 2.39 (s, NCH3), 3.23 (dd,
J ¼ 4:2, 7.2 Hz, CH), 3.35 (s, OCH3), 3.53 (dd, J ¼ 7:2,
9.6 Hz, CHH0OCH3), 3.67 (dd, J ¼ 4:2, 9.6 Hz,
CHH0OCH3), 4.42 (dd, J ¼ 5:5, 14.4 Hz, CHH0Ph), 4.49
(dd, J ¼ 5:9, 14.4 Hz, CHH0Ph), 7.23–7.36 (m, 5PhH),
7.73 (br s, NH); 13C NMR (CDCl3) 35.3, 42.9, 58.7,
64.6, 72.3, 127.2, 127.4, 128.5, 138.3, 171.6 ppm; MS
(+CI) (rel intensity) 224 (13), 223 (Mþ+1, 100), 102 (73),
101 (23), 100 (32); Mr (+CI) 223.14436 [Mþ+1] (calcd for
Compound 28 (1.32 g, 67%) was prepared as a colorless
oil from 45 (2.03 g, 7.40 mmol) and 2-methoxyethyl-
amine (20 mL, 230 mmol) at 45 ꢁC (2 d) utilizing method
A. Purification was by column chromatography (SiO2;
1:49, MeOH–CHCl3): Rf 0.28 (1:49, MeOH–CHCl3); IR
(neat) 3321 (br), 1662, 1524, 700 cmÀ1
;
1H NMR
(CDCl3) d 2.24 (br s, NH), 2.62–2.86 (m, NCH2), 3.24
C12H19N2O2
223.14465).
Anal.
Calcd
for

3094
C. Bꢀeguin et al. / Bioorg. Med. Chem. 12 (2004) 3079–3096
(s, OCH3), 3.31–3.40 (m, CH, OCH3, and CH2OCH3),
3.52–3.57 (m, CHCHH0OCH3), 3.62–3.67 (m,
CHCHH0OCH3), 4.42 (d, J ¼ 6:0 Hz, CH2Ph), 7.22–
7.32 (m, 5PhH), 7.96 (br s, NH); 13C NMR (CDCl3)
42.5, 47.6, 58.1, 58.2, 62.1, 71.4, 72.3, 126.7, 127.0,
128.0, 138.0, 171.5 ppm; MS (+CI) (rel intensity) 268
(17), 267 (Mþ+1, 100), 208 (26); Mr (+CI) 267.17120
[Mþ+1] (calcd for C14H23N2O3 267.17087).
method A. Purification was by column chromatography
(SiO2; 1:49, MeOH–CHCl3): mp 85–86 ꢁC; Rf 0.29 (1:49,
MeOH–CHCl3); IR (KBr) 3228, 3042, 2925, 2183, 2044,
1645, 1544, 1242, 1090, 971, 698 cmÀ1
;
1H NMR
(CDCl3) d 3.76 (s, CH), 4.43 (dd, J ¼ 5:8, 14.4 Hz,
CHH0Ph), 4.49 (dd, J ¼ 6:1, 14.4 Hz, CHH0Ph), 7.21–
7.38 (m, 10PhH), the signal for the NH proton was not
detected; 13C NMR (CDCl3) 43.2, 77.0, 127.4, 127.8,
128.0, 128.4, 128.6, 128.7, 136.6, 138.4, 171.6 ppm, the
CD3 signals were not detected; MS (+CI) (rel intensity)
276 (15), 275 (Mþ+1, 100), 140 (24); Mr (+CI) 275.20381
[Mþ+1] (calcd for C17H15D6N2O 275.20305). Anal.
Calcd for C17H14D6N2O: C, 74.61; H, 7.35 (calculated
H+D as H); N, 10.19. Found: C, 74.61; H, 7.41; N,
10.19.
4.22. Synthesis of( R,S)-N-benzyl-2-dimethylamino-3-
methoxypropionamide (29)
Compound 29 (3.15 g, 97%) was prepared as a pale-
yellow oil from 45 (3.75 g, 13.79 mmol) and dimethyl-
amine (2.0 M in THF, 25 mL, 50 mmol) at room tem-
perature (18 h) utilizing method A. Purification was by
column chromatography (SiO2; 1:49, MeOH–CHCl3):
Rf 0.27 (1:49, MeOH–CHCl3); IR (neat) 3322 (br), 1662,
4.25. Synthesis of( R)-N-benzyl-2-ethylamino-3-meth-
oxypropionamide ((R)-26)
1
1523 cmÀ1; H NMR (CDCl3) d 2.35 (s, N(CH3)2), 3.05
(dd, J ¼ 3:6, 5.7 Hz, CH), 3.24 (s, OCH3), 3.71 (dd,
J ¼ 5:7, 10.5 Hz, CHH0OCH3), 3.82 (dd, J ¼ 3:6,
10.5 Hz, CHH0OCH3), 4.43 (dd, J ¼ 5:3, 14.4 Hz,
CHH0Ph), 4.50 (dd, J ¼ 5:3, 14.4 Hz, CHH0Ph), 7.22–
7.35 (m, 5PhH), 7.57 (br s, NH); 13C NMR (CDCl3)
42.8, 43.0, 58.8, 69.6, 70.4, 127.2, 127.5, 128.5, 138.5,
171.4 ppm; MS (+CI) (rel intensity) 238 (17),
237 (Mþ+1, 100), 102 (41); Mr (+CI) 237.15994 [Mþ+1]
(calcd for C13H21N2O2 237.16030). Anal. Calcd for
C13H20N2O2: C, 66.07; H, 8.53; N, 11.86. Found: C,
65.89; H, 8.69; N, 11.81.
To a MeOH (50 mL) solution of (R)-246 (887 mg,
4.24 mmol) and sodium cyanoborohydride (187 mg,
2.97 mmol) was added acetaldehyde (250 lL, 4.45 mmol)
in one portion and then the mixture was stirred at room
temperature (10 min). The reaction solvent was evapo-
rated and the crude was purified by column chroma-
tography (SiO2; 1:49, MeOH–CHCl3) to obtain (R)-26
as a pale-yellow oil (511 mg, 51%): Rf 0.32 (1:49,
MeOH–CHCl3); IR (neat) 3319 (br), 3062, 3033, 1661,
1526 cmÀ1
;
1H NMR (CDCl3) d 1.08 (t, J ¼ 7:2 Hz,
CH2CH3), 1.64 (br s, NH), 2.55–2.71 (m, CH2CH3),
3.31–3.37 (m, CH), 3.36 (s, OCH3), 3.56 (dd, J ¼ 7:2,
9.6 Hz, OCHH0), 3.67 (dd, J ¼ 4:2, 9.6 Hz, OCHH0),
4.47 (d, J ¼ 6:3 Hz, CH2Ph), 7.25–7.36 (m, 5PhH), 7.84
(br s, NH), addition of excess (R)-())-mandelic acid27 to
a CDCl3 solution containing (R)-26 gave only one signal
for the methoxy methyl protons (d 3.06) and the N-ethyl
methyl protons (d 1.02); 13C NMR (CDCl3) 15.3, 42.9,
43.1, 58.7, 62.5, 72.6, 127.2, 127.3, 128.5, 138.3,
172.1 ppm; MS (+CI) (rel intensity) 238 (15), 237
(Mþ+1, 100), 102 (15); Mr (+CI) 237.16032 [Mþ+1]
(calcd for C13H21N2O2 237.16030). Anal. Calcd for
C13H20N2O2Æ0.25H2O: C, 64.84; H, 8.58; N, 11.63.
Found: C, 64.89; H, 8.60; N, 11.59.
4.23. Synthesis of( R,S)-N-benzyl-2-morpholin-4-yl-3-
methoxypropionamide (30)
Compound 30 (950 mg, 80%) was prepared as a clear
thick oil from 45 (1.17 g, 4.30 mmol) and morpholine
(20 mL, 229 mmol) at 45 ꢁC (2 d) utilizing method A.
Purification was by column chromatography (SiO2;
1:49, MeOH–CHCl3): Rf 0.35 (1:49, MeOH–CHCl3); IR
1
(neat) 3319 (br), 1662, 1521 cmÀ1; H NMR (CDCl3) d
2.54–2.73 (m, CH2NCH2), 3.11 (dd, J ¼ 3:5, 5.6 Hz,
CH), 3.32 (s, OCH3), 3.60–3.74 (m, CH2OCH2 and
CHH0OCH3), 3.83 (dd, J ¼ 3:5, 10.4 Hz, CHH0OCH3),
4.42 (dd, J ¼ 6:0, 14.9 Hz, CHH0Ph), 4.50 (dd, J ¼ 6:3,
14.9 Hz, CHH0Ph), 7.24–7.35 (m, 5PhH), 7.63 (br s,
NH); 13C NMR (CDCl3) 42.9, 50.8, 58.7, 66.9, 68.7,
69.6, 127.1, 127.2, 128.4, 138.2, 170.6 ppm; MS (+CI)
(rel intensity) 280 (16), 279 (Mþ+1, 100), 144 (11); Mr
(+CI) 279.17118 [Mþ+1] (calcd for C15H23N2O3
279.17087). Anal. Calcd for C15H22N2O3Æ0.2H2O: C,
63.90; H, 8.01; N, 9.94. Found: C, 63.81; H, 8.06; N,
10.00.
4.26. Synthesis of( R,S)-N-benzyl-2-dimethylamino-2-
phenylacetamide-N-oxide (51)
A CH2Cl2 (3 mL) solution of 19 (50 mg, 0.19 mmol) and
m-chloroperoxybenzoic acid (42 mg, 0.19 mmol) was
stirred at room temperature (2 h). The solvent was
evaporated in vacuo and the residue was purified by
PTLC (SiO2, 1:1, Et2O–MeOH): Rf 0.27 (1:1, Et2O–
1
MeOH); H NMR (CDCl3) d 2.89 (s, NCH3), 3.32 (s,
4.24. Synthesis of( R,S)-N-benzyl-2-dimethyl-d6-amino-2-
phenylacetamide (46)
NCH03), 4.50 (dd, J ¼ 5:9, 15.3 Hz, CHH0Ph), 4.56 (dd,
J ¼ 6:4, 15.3 Hz, CHH0Ph), 4.75 (s, CH), 11.71 (br s,
NH), 7.24–7.43 (m, 8PhH), 7.70 (d, J ¼ 7:5 Hz, 2PhH),
11.71 (br s, NH); 13C NMR (CDCl3) 42.9, 58.0, 59.6,
80.9, 127.1, 127.6, 128.3, 128.6, 130.0, 130.9, 138.1,
166.6 ppm, one aromatic carbon signal was not detected
and is believed to overlap with nearby peaks.
Compound 46 (710 mg, 55%) was prepared as a white
solid from 39 (1.22 g, 4.70 mmol), dimethyl-d6-amine
hydrochloride (5.00 g, 57.08 mmol), Et3N (8.0 mL,
57.08 mmol), and THF (20 mL) at 50 ꢁC (7 d) utilizing

C. Bꢀeguin et al. / Bioorg. Med. Chem. 12 (2004) 3079–3096
3095
4.27. Pharmacology
4.30. LC–MS analysis ofmetabolic products
Compounds were screened under the auspices of the
National Institutes of Health’s Anticonvulsant Screen-
ing Project. Experiments were performed in male
rodents [albino Carworth Farms No. 1 mice (intraperi-
toneal route, ip), albino Sprague-Dawley rats (oral
route, po)]. Housing, handling, and feeding were all in
accordance with recommendations contained in the
‘Guide for the Care and Use of Laboratory Animals’.
Anticonvulsant activity was established using the MES
test,7;16 the scMet test,7 and the timed intravenous
pentylenetetrazole (iv Metrazol) seizure threshold test,7
using previously reported methods.1l;10
The metabolic profiles were analyzed using LC–MS/MS
on an Agilent 1100 LC/MSD Trap system at LCMS
Limited (Raleigh, NC). Samples (5 lL) were injected
onto a Xterra Phenyl column (2.0 · 150 mm, Waters
Corp., Catalog No. 186001181). A gradient mobile
phase was employed and was first held for 4 min at 80:20
ratio of solvent A (10 mM ammonium acetate buffer
adjusted to pH 5.5) and solvent B (MeCN) and then
successively brought to a 40:60 A–B composition over a
10 min period and then a 20:80 A–B composition over a
2 min interval using a flow rate of 0.2 mL/min. A 3 min
recycling period was then employed. Positive ion elec-
trospray over the range m=z 70–400 was used for
detection. Some of the key electrospray MS operating
conditions are as follows: capillary voltage, 3000 V;
capillary exit voltage, 60 V; trap drive, m=z 43; drying
gas temperature, 350 ꢁC; drying gas flow rate, 9.0 L/min;
nebulization pressure, 45 psi. The ion trap was also
operated in auto-MS/MS mode to gain structural
information on the metabolites. Parent ion isolated
using the auto-MS/MS routine selects the base peak in
the mass spectrum for collision-induced decomposition
(CID), corresponding to the [M+H]þ ion for the
metabolites in this paper. The product ions were gen-
erated by CID of the parent ion with He at a frag-
mentation voltage linear ramp from 0.3–2 V over 30 ms
duration and a cutoff mass of 35% of the parent ion m=z
value. The product ion spectra were acquired from m=z
70 to 400.
4.28. Microsomal studies. General methods
Male rat (Catalog No. 452501) and mouse (Catalog No.
452701) liver microsomes were purchased from GEN-
TEST Corp (Woburn, MA). b-Nicotinamide adenine
dinucleotide phosphate reduced (NADPH) was pur-
chased from Sigma/RBI (St. Louis, MO). All other
reagents were of best commercial grade available and
were used without further purification. Microsomal
incubations were performed at room temperature in
Corning centrifuge tubes containing the substrate
(100 lM), NADPH (1 mM), the buffer (100 mM phos-
phate buffer at pH 7.4 containing 3.3 mM MgCl2), and
the rodent liver microsomal protein (1 mg/mL) in a total
volume of 2 mL. The substrates were dissolved in a
MeOH in such concentrations that the total organic
solvent content did not exceed 1% when added to
microsomal incubations. The total incubation time was
1 h. Incubations in the absence of NADPH were used as
controls, and incubations without any substrate were
used as blanks. Samples (300 lL) were taken at 15 min
interval points into Eppendorf tubes containing MeCN
(150 lL). The incubation mixtures were then centrifuged
for 10 min (14,000 rpm) in an Eppendorf centrifuge
5415C. An aliquot of the supernatant was analyzed by
HPLC–UV and HPLC–MS/MS.
Acknowledgements
The authors thank the NINDS and the Anticonvulsant
Screening Project (ASP) at the National Institutes of
Health for kindly performing the pharmacological
studies via the ASP’s contract site at the University of
Utah with Drs. H. Wolfe, S. White, and K. Wilcox. We
thank Drs. John Ansede, James E. Hall, and Dhiren
Thakker (University of North Carolina-Chapel Hill) for
their advice and help in conducting the microsomal
studies.
4.29. LC–UV analysis ofmetabolic products
To determine the substrate metabolic profile, the
microsomal incubations at various time intervals were
analyzed by LC–UV (256 nm). A Waters Alliance 2690
HPLC was interfaced with a photodiode array detector
(Waters Photodiode Array, model 996). Samples (50 lL)
were injected onto an Xterra Phenyl column
(3.9 · 150 mm, Waters Corp., Catalog No. 186001184).
A gradient mobile phase was employed and was first
held for 4 min at 80:20 ratio of solvent A (10 mM
ammonium acetate buffer adjusted to pH 5.5) and sol-
vent B (MeCN) and then successively brought to a 40:60
A–B composition over a 10 min period and then a 20:80
A–B composition over a 2 min interval using a flow rate
of 0.8 mL/min. A 3 min recycling period was then
employed. Calibration curves using compounds 14, 15,
16, 19, 20, and 32 were used for the quantification of the
substrates and major metabolites.
References and notes
1. (a) Cortes, S.; Liao, Z.-K.; Watson, D.; Kohn, H. J. Med.
Chem. 1985, 28, 601; (b) Conley, J. D.; Kohn, H. J. Med.
Chem. 1987, 30, 567; (c) Kohn, H.; Conley, J. D. Chem.
Br. 1988, 24, 231; (d) Kohn, H.; Conley, J. D.; Leander, J.
D. Brain Res. 1988, 457, 371; (e) Kohn, H.; Sawhney, K.
N.; LeGall, P.; Conley, J. D.; Robertson, D. W.; Leander,
J. D. J. Med. Chem. 1990, 33, 919; (f) Kohn, H.; Sawhney,
K. N.; LeGall, P.; Robertson, D. W.; Leander, J. D. J.
Med. Chem. 1991, 34, 2444; (g) Kohn, H.; Sawhney, K.
N.; Bardel, P.; Robertson, D. W.; Leander, J. D. J. Med.
Chem. 1993, 36, 3350; (h) Bardel, P.; Bolanos, A.; Kohn,
H. J. Med. Chem. 1994, 37, 4567; (i) Kohn, H.; Sawhney,
K. N.; Robertson, D. W.; Leander, J. D. J. Pharm. Sci.
1994, 83, 689; (j) Choi, D.; Stables, J. P.; Kohn, H. J. Med.
Chem. 1996, 39, 1907; (k) Shen, M.; LeTiran, A.; Xiao, Y.;

3096
C. Bꢀeguin et al. / Bioorg. Med. Chem. 12 (2004) 3079–3096
Golbraikh, A.; Kohn, H.; Tropsha, A. J. Med. Chem. 12. Hoffman, R. V.; Nayyar, N. K.; Chen, W. T. J. Org.
2002, 45, 2811; (l) LeTiran, A.; Stables, J. P.; Kohn, H.
J. Med. Chem. 2002, 45, 4762.
Chem. 1993, 58, 2355.
13. Shah, K. J.; Trivedi, J. J. Indian J. Appl. Chem. 1967, 11.
14. Brain, C. T.; Hallett, A.; Ko, S. Y. J. Org. Chem. 1997, 62,
3808.
15. Jong-Man, K.; Wilson, T. E.; Norman, T. C.; Schultz,
P. G. Tetrahedron Lett. 1996, 37, 5309.
16. Krall, R. L.; Penry, J. K.; White, B. G.; Kupferberg, H. J.;
Swinyard, E. A. Epilepsia 1978, 19, 409.
17. Porter, R. J.; Cereghino, J. J.; Gladding, G. D.; Hessie,
B. J.; Kupferberg, H. J.; Scoville, B.; White, B. G.
Cleveland Clin. Q. 1984, 51, 293.
2. (a) Paruszewski, R.; Rostafinska-Suchar, G.; Strupinska,
M.; Jaworski, P.; Stables, J. P. Pharmazie 1996, 3, 145; (b)
Paruszewski, R.; Rostafinska-Suchar, G.; Strupinska, M.;
Jaworski, P.; Winiecka, I.; Stables, J. P. Pharmazie 1996,
51, 212; (c) Paruszewski, R.; Rostafinska-Suchar, G.;
Strupinska, M.; Winiecka, I.; Stables, J. P. Pharmazie
2000, 55, 27; (d) Paruszewski, R.; Strupinska, M.; Stables,
J. P.; Swiader, M.; Czuczwar, S.; Kleinrock, Z.; Turski, W.
Chem. Pharm. Bull. 2001, 49, 629.
3. Ho, B.; Venkatarangan, P. M.; Cruse, S. F.; Hinko, C. N.;
Andersen, P. H.; Crider, A. M.; Adloo, A. A.; Roane, D.
S.; Stables, J. P. Eur. J. Med. Chem. 1998, 33, 23.
4. Andurkar, S. V.; Stables, J. P.; Kohn, H. Bioorg. Med.
Chem. 1999, 7, 2381.
18. (a) Bialer, M.; Johannessen, S. I.; Kupferberg, H. J.; Levy,
R. H.; Loiseau, P.; Perucca, E. Epilepsy Res. 2001, 43, 11;
(b) Stables, J. P.; Kohn, H. Unpublished results.
19. A similar response was observed for 17 when administered
to mice (ip) at 46 and 83 mg/kg.
5. Choi, D.; Stables, J. P.; Kohn, H. Bioorg. Med. Chem.
1996, 4, 2105.
20. Stables, J. P. Private communication based on unpub-
lished data performed by the ASP.
6. Andurkar, S. V.; Stables, J. P.; Kohn, H. Tetrahedron:
Asymmetry 1998, 9, 3841.
7. Stables, J. P.; Kupferberg, H. J. In Molecular and Cellular
Targets for Antiepileptic Drugs; Avanzini, G., Tanganelli,
P., Avoli, M., Eds.; John Libbey: London, 1997; pp 191–
198.
8. White, H. S.; Woodhead, J. H.; Franklin, M. R. In
Antiepileptic Drugs; Levy, R. H., Mattson, R. H.,
Meldrum, B. S., Eds.; 4th ed.; Raven: New York, 1995;
pp 99–110.
9. Dunham, M. S.; Miya, T. A. J. Am. Pharm. Assoc. Sci.
Edit. 1957, 46, 208.
21. Surh, Y.-J.; Ahn, S. H.; Kim, K.-C.; Park, J.-B.; Sohn,
Y. W.; Lee, S. S. Life Sci. 1995, 56, 305, and references
cited therein.
22. Rose, J.; Castagnoli, N., Jr. Med. Res. Rev. 1983, 3, 73.
23. Bickel, M. H. Arch. Biochem. Biophys. 1972, 148, 54.
24. Beckett, A. H.; Triggs, E. J. Nature 1966, 211,
1415.
25. Guengerich, F. P.; Yun, C. H.; Macdonald, T. L. J. Biol.
Chem. 1996, 271, 27321.
26. Kushner, S.; Cassell, R. I.; Morton, J., II; Williams, J. H.
J. Org. Chem. 1951, 16, 1283.
27. For comparable procedures for resolving stereoisomers,
see the following. (a) Weisman, G. R. In Asymmetric
Synthesis––Analytical Methods; Morrison, J. D., Ed.;
Academic: New York, 1983; Vol. 1, pp 153–171; (b)
Parker, D.; Taylor, R. J. Tetrahedron 1987, 43, 5431.
10. LeTiran, A.; Stables, J. P.; Kohn, H. Bioorg. Med. Chem.
2001, 9, 2693.
11. Carter, H. E.; West, D. W. Org. Synth. Coll. Vol. III, 1955;
pp 774–775.